Effects of Dietary Carbohydrate
and Fat Intake on Glucose and
Lipoprotein Metabolism in
Individuals With Diabetes Mellitus
The dietary treatment of individuals with diabetes
remains a controversial issue. The major emphasis
in recent years has been on the reduction of total
fat and saturated fat and replacement with complex
carbohydrate. The rationale for this approach is based
on the premise that such diets will reduce the risk of
coronary artery disease (CAD) by reducing total and lowdensity lipoprotein cholesterol concentrations. In this
article, we review the available data and conclude that
there is little evidence to support the notion that low-fat
high-carbohydrate diets per se lead to any reduction in
the risk for CAD in individuals with diabetes. The only
data indicating that low-fat high-carbohydrate diets lead
to beneficial effects on carbohydrate and lipoprotein
metabolism are confounded either by the lack of
suitable experimental control, by the fact that diets also
differed in the type of dietary fat and amount of dietary
cholesterol, or were enormously enriched in dietary
fiber. When these factors are taken into consideration,
there appears to be little evidence in support of the
view that substituting carbohydrate for fat in the diets
of individuals with diabetes results in any measurable
beneficial effect. Indeed, it could be argued that
the most characteristic defects in carbohydrate and
lipoprotein metabolism are exacerbated in response
to low-fat high-carbohydrate diets. Alternatively,
the data presented herein strongly suggest that diets
containing conventional quantities of fat, in which
saturated fat is replaced by unsaturated fat and dietary
cholesterol reduced, would result in the desired
reductions to total and low-density lipoprotein
cholesterol concentrations without the adverse
effects of increased postprandial glucose and insulin
From the Department of Medicine and General Clinical Research Center, Stanford University School of Medicine, Stanford; and the Geriatric Research, Education, and Clinical Center, Veterans Administration Medical Center, Palo
Alto, California.
Address correspondence and reprint requests to Clarie B. Hollenbeck, PhD,
GCRC HG323, Stanford University Medical Center, Stanford, CA 94305.
774
Clarie B. Hollenbeck, PhD
Ann M. Coulston, MS, RD
concentrations, increased fasting and postprandial
total and very-low-density lipoprotein triglyceride
concentrations, and decreased fasting high-density
lipoprotein cholesterol concentrations. Diabetes Care
14:774-85, 1991
A
lthough there is general agreement that diet
should serve as the cornerstone in the treatment
of individuals with diabetes, the exact nature of
the diet remains controversial (1-4). Interest in
dietary management has intensified in recent years, associated with heightened efforts to normalize carbohydrate and lipoprotein metabolism in individuals with
diabetes. However, an understanding of the impact that
differences in dietary composition have on the metabolic abnormalities present within this syndrome is far
from clear. In this review, we attempt to clarify some of
the dietary issues related to the treatment of carbohydrate intolerance and dyslipemia present in the diabetic
syndrome. We will draw our comments from our experience (5-10) over the past several years, as well as
the work of others (11-23). Furthermore, we limit our
comments to studies that have assessed the effects of
dietary changes under controlled well-defined dietary
conditions.
Several studies have been conducted in outpatient
settings where individuals are given dietary guidelines,
asked to obtain specific foods, and follow the advice for
periods up to 6 wk (24-26). Because the experimental
variable in these studies (i.e., diet) is outside the control
of the investigator, it is difficult to accept these data as
persuasive scientific evidence of the effects that specific
dietary changes might have on the regulation of carbohydrate and lipoprotein metabolism. Unfortunately,
DIABETES CARE, VOL . 14, N O . 9, SEPTEMBER 1991
C.B. HOLLENBECK AND A.M. COULSTON
many of these latter studies play a prominent role in the
rationale for the dietary advice being given (1,2).
Finally, before turning our attention to an analyses
that specific dietary manipulation might have on carbohydrate and lipoprotein metabolism, it would seem
useful to consider the differences in carbohydrate and
lipoprotein metabolism that characterize individuals
with diabetes and the relationship of these abnormalities
to the risk of coronary artery disease (CAD).
ABNORMALITIES OF CARBOHYDRATE AND LIPOPROTEIN
METABOLISM IN DIABETES
Abnormalities in carbohydrate metabolism are present
in patients with both insulin-dependent diabetes mellitus (IDDM) and non-insulin-dependent diabetes mellitus
(NIDDM) (27,28). However, the nature of these abnormalities vary considerably between the two syndromes.
Individuals with IDDM are characterized by the total
absence of insulin secretion as a result of pancreatic (3cell destruction (27). Thus, the primary cause of hyperglycemia in IDDM is related to insulin deficiency. On
the other hand, although individuals with NIDDM may
have decreased insulin secretion in response to a glucose stimulus, most individuals maintain normal or
greater than normal levels of circulating insulin in response to mixed meals (28-30). The primary cause
of ambient hyperglycemia in these individuals appears to be reduced insulin sensitivity of peripheral
tissue rather than reduced insulin secretion (28). Consequently, insulin resistance and compensatory hyperinsulinism represent the major metabolic abnormality in
carbohydrate metabolism present in NIDDM.
Although diabetes is usually characterized as a disease of carbohydrate metabolism, abnormalities of lipoprotein metabolism are also present in individuals
with this syndrome. The abnormalities are particularly
important given the acceleration of CAD associated with
diabetes and the public and scientific attention given to
the possible relationship between diet and CAD. The
dyslipemia present in diabetes also appears to vary considerably between the two syndromes. For example,
hypertriglyceridemia is common in individuals with uncontrolled IDDM. This hypertriglyceridemia is almost
certainly due to a defect in the removal of triglyceriderich lipoprotein chylomicrons and very-low-density lipoproteins (VLDLs) from plasma (31). Hypertriglyceridemia is also common in individuals with NIDDM;
however, the elevated plasma triglyceride concentrations appear to result primarily from increased endogenous triglyceride synthesis (32-35). Moreover, although plasma triglyceride concentrations may fall in
response to improved glycemic control in both syndromes, the reasons for the reduction and the degree to
which normal plasma triglyceride concentrations are
achieved differ in IDDM compared with NIDDM. The
defect in triglyceride removal in IDDM seems to be related to the absence of circulating plasma insulin. In-
DIABETES CARE, V O L . 14, N O . 9, SEPTEMBER 1991
sulin-replacement therapy appears to correct the defect
in the removal rate of chylomicrons and VLDL triglyceride in individuals with IDDM, and plasma triglyceride
concentrations can be lowered to relatively normal levels, although they may remain hyperglycemic (34). In
contrast, the reduction in plasma triglyceride concentrations in individuals with NIDDM is associated with
improved glycemic control and is probably related to a
reduction in VLDL triglyceride secretion, secondary to
a fall in free fatty acid (FFA) concentrations, which occur
with improvements in glycemic control (33,34). Furthermore, individuals with NIDDM continue to manifest
an increase in VLDL triglyceride secretion rates and hypertriglyceridemia compared with individuals with
IDDM matched for plasma glucose and FFA concentrations (34).
In addition to the differences in VLDL metabolism
noted between individuals with IDDM and NIDDM,
there also seems to be differences in high-density lipoprotein (HDL) metabolism. For example, HDL cholesterol concentrations are generally not reduced and may
even be increased in individuals with IDDM (36). In
contrast, low plasma HDL cholesterol concentrations
occur frequently in individuals with NIDDM (37). Note
that in both IDDM and NIDDM, plasma low-density
lipoprotein (LDL) cholesterol concentrations do not differ substantially from individuals without diabetes (38).
This observation is particularly interesting, because the
major thrust of dietary advice for patients with diabetes
is aimed at lowering plasma LDL cholesterol concentrations.
At this point in our discussion it would seem important to relate these abnormalities in carbohydrate and
lipoprotein metabolism to the risk of developing CAD
in the diabetic population. In this regard, hypertriglyceridemia has been associated with an increased incidence of CAD in both cross-sectional and prospective
studies (39-42). In the World Health Organization multinational study (41), the single risk factor that correlated
the strongest with the incidence of CAD in the nine
diabetic populations studied was plasma triglyceride
concentrations. Similar evidence of the importance of
plasma triglyceride as a risk factor for the development
of CAD in individuals with diabetes is provided by the
Paris Prospective Study (42). In this study, although
plasma triglyceride, cholesterol, and insulin all correlated with the incidence of CAD by univariate analysis,
plasma triglyceride was the only factor that was significantly associated with CAD by multivariate regression
analysis. The results of these two studies serve to underscore the importance of hypertriglyceridemia as a
risk factor for CAD in individuals with diabetes. Unfortunately, these two studies assessed only total plasma
triglyceride and cholesterol concentrations and did not
evaluate various lipoprotein fractions (41,42). In addition to hypertriglyceridemia, low HDL cholesterol concentrations have been shown to be associated with an
increased risk for CAD in individuals without diabetes
(43,44), and there is evidence that the reduced HDL
775
GLUCOSE AND LIPOPROTEIN METABOLISM IN DIABETES
cholesterol levels seen in individuals with NIDDM also
contribute to enhanced atherosclerosis observed within
this syndrome (45,46). There appears to be a close inverse relationship between VLDL triglyceride and HDL
cholesterol (47,48). The decline in HDL cholesterol,
which accompanies the increase in VLDL triglyceride,
has been suggested to reflect the interchange of cholesterol esters and triglyceride between HDL and VLDL
particles, representing a recycling of cholesterol esters
from HDL back to VLDL (48).
Although increased concentrations of both total
plasma and LDL cholesterol are strong risk factors for
CAD in the general population, they have not been
identified as particularly strong risk factors for CAD in
individuals with diabetes. As mentioned previously, total and LDL cholesterol are perhaps the most normal
aspect of lipoprotein metabolism in diabetes and do not
appear to be elevated compared with a normal glucosetolerant population.
Finally, the idea that elevated plasma insulin concentrations may play a role in the development of CAD was
proposed >20 yr ago (49). Since that time, evidence in
support of that hypothesis has accumulated in various
patient populations (50-52), and was extensively reviewed recently (53). Hyperinsulinemia is a common
finding in individuals with NIDDM (30,40,42), and
there is evidence that hyperinsulinemia may play an
important role in the development of CAD in this syndrome as well (40,42,50).
Obviously, a major evaluation of the effects of dietary
manipulation on carbohydrate and lipoprotein metabolism must take into consideration the relationship between any observed changes in metabolism and the risk
for CAD in individuals with diabetes. In particular, attention must be directed to the impact that specific dietary changes have on the metabolic risk factors for CAD
that are known to exist in individuals with diabetes.
These dietary changes have been outlined above and
are the focus of our discussion on the effects of diet on
carbohydrate and lipoprotein metabolism in this review.
ISOCALORIC SUBSTITUTION OF CARBOHYDRATE FOR FAT
Normal glucose-tolerant individuals. Several years
ago, we demonstrated that the isocaloric substitution of
carbohydrate for fat in the diets of individuals with normal carbohydrate and lipoprotein metabolism resulted
in a significant increase in fasting total plasma triglyceride ( + 49%, P < 0.001) and a significant reduction in
HDL cholesterol (-18%, P < 0.01) (5). Of particular
interest is the observation that deleterious effects to lipoprotein metabolism occurred in the absence of the
anticipated improvement in total plasma or LDL cholesterol concentrations. In addition, postprandial triglyceride and insulin concentrations were significantly
elevated after the low-fat diet. In the same year we reported a significant increase in fasting triglyceride concentrations (238 vs. 348 mg/dl, P < 0.001) in a group
776
of nondiabetic individuals with endogenous hypertriglyceridemia when carbohydrate was substituted for
fat in the diet (6). The increase in triglyceride could
be explained primarily by a significant increase in
VLDL triglyceride concentration (147 vs. 241 mg/dl,
P < 0.001). Again, the expected changes in total plasma
and LDL cholesterol were not observed on the low-fat
diet. In addition, as was observed in the nondiabetic
individuals cited above (5), postprandial insulin concentrations were significantly increased when carbohydrate was substituted for fat in the diet. Grundy (11)
and Grundy et al. (12) reported similar findings when
saturated fat was replaced by either complex carbohydrate or monounsaturated fat in the diets of individuals
with normal carbohydrate and lipoprotein metabolism.
In their first study, these investigators demonstrated a
significant increase in total plasma cholesterol (208 vs.
222 mg/dl, P < 0.05) and triglyceride concentrations
(178 vs. 235 mg/dl, P < 0.05) after the high-carbohydrate low-fat diet (11). In addition, HDL cholesterol was
significantly decreased (39 vs. 32 mg/dl, P < 0.01).
Because no significant change in LDL cholesterol concentration was observed, the ratio of LDL to HDL cholesterol was significantly increased (3.5 vs. 4.7, P <
0.01) after the low-fat diet. Their second study also
showed significant decrease in HDL cholesterol concentrations (50 vs. 46 mg/dl, P < 0.02) without any
significant change in total plasma or LDL cholesterol
concentrations after the high-carbohydrate diet (12).
Thus, there now appears to be substantial evidence
in individuals without diabetes, which suggests that lowfat diets per se do not provide any significant advantage
in lowering total plasma or LDL cholesterol concentrations than diets in which saturated fats are replaced with
unsaturated fats (either monounsaturated or polyunsaturated). More importantly, diets increased in carbohydrates have a tendency to result in significantly higher
fasting and postprandial triglyceride concentrations, increased postprandial insulin concentrations, and lower
HDL cholesterol concentrations, all of which have been
associated with an increased risk of CAD. The significant fall in HDL cholesterol in the absence of any
change in total or LDL cholesterol concentrations appears to be of substantial clinical importance. With the
data on nondiabetic individuals as a background, we
would like to turn our attention to individuals with diabetes.
Individuals with diabetes mellitus. The effects of isocaloric substitution of carbohydrate for fat in the diets
of individuals with diabetes appear to be similar to individuals with normal glucose tolerance. Studies conducted by our group (7,8) and others (13-15) have
consistently indicated that the substitution of carbohydrate for fat in the diet of individuals with diabetes results in an increase in plasma glucose, insulin, total
plasma and VLDL triglyceride concentrations, and a reduction in HDL cholesterol concentrations without appreciably affecting total plasma or LDL cholesterol
concentrations. For example, two studies by our group
DIABETES CARE, VOL . 14, N O . 9, SEPTEMBER 1991
C.B. HOLLENBECK AND A.M. COULSTON
showed that the isocaloric substitution of carbohydrate
for fat resulted in significant increases in fasting and
postprandial total plasma triglyceride concentrations
(7,8; Fig. 1). The increase in fasting total triglyceride
concentrations could be explained almost completely
by a significant increase in VLDL triglyceride concentrations (Fig. 2). In both studies, there was also a significant increase in VLDL cholesterol and a significant
decrease in HDL cholesterol concentrations after the
low-fat diet (Fig. 3). The only difference in lipoprotein
response between the two studies was that although in
one study (7) no significant improvement in LDL cholesterol concentration was observed, there was a small
but significant fall in LDL cholesterol after the low-fat
diet in the other (8). In addition to the changes in lipoprotein metabolism in these studies (7,8), there were
also alterations in carbohydrate metabolism observed.
Specifically, postprandial plasma glucose and insulin
concentrations were significantly elevated after the highcarbohydrate diets on both studies (Fig. 4). Further evidence of deterioration in glycemic control is apparent
from the significant increase in 24-h urinary glucose excretion after the high-carbohydrate diet. Similar results
have been reported by Sestoft et al. (13). These investigators demonstrated that increasing dietary carbohydrate as polysaccharide from 40 to 50% of total calories
in individuals with NIDDM resulted in a significant increase in total plasma and VLDL triglyceride concentrations and a significant decrease in HDL cholesterol
concentrations. In addition, there was a significant in-
8
9
10
II
12
1
2
3
4
Clock Hours
• ^ 40% CHO Diet
o-o 60% CHO Diet
FIG. 1. An example of fasting and postprandial plasma
triglyceride concentrations in individuals with non-insulindependent diabetes after diets containing either 40% (•)
or 60% (o) of total calories as carbohydrate (CHO). From
Coulston et al. (7). © by the American journal of Medicine.
DIABETES CARE, VOL. 14, N O . 9, SEPTEMBER 1991
250
Triglyceride
(mg/dl)
p<0.001
VLDL-TG
(mg/dl)
p<0.001
200
150
I
I
100
50
D40%
060%
CHO Diet
CHO Diet
FIG. 2. Elevation in total plasma triglyceride (TG) concentrations frequently observed in individuals with non-insulin-dependent diabetes after high-carbohydrate (CHO)
low-fat diets can be explained by significant increase in
very-low-density lipoprotein (VLDL) TG. From Coulston et
al. (7). © by the American journal of Medicine.
crease in both postprandial glucose and insulin concentrations after the high-carbohydrate low-fat diet.
Finally, Garg et al. (14) demonstrated that replacing
saturated fats with monounsaturated fats resulted in significant metabolic advantages over replacing saturated
fat with complex carbohydrate. Specifically, there were
no significant changes in either fasting total plasma or
LDL cholesterol concentrations as a result of reducing
total fat from 50 to 25% of the total calories. However,
fasting total plasma triglyceride (163 vs. 218 mg/dl, P <
0.01) and VLDL cholesterol concentrations (28 vs. 43
mg/dl, P < 0.005) and total-HDL cholesterol ratio (6
vs. 7.2, P < 0.05) were significantly increased and HDL
cholesterol concentrations (34 vs. 30 mg/dl, P < 0.005)
significantly decreased after the high-carbohydrate lowfat diet. In addition, postprandial plasma glucose
(+12%, P < 0.01), insulin ( + 24%, P < 0.01), and
triglyceride ( + 26%, P < 0.005) concentrations were all
significantly elevated after the high-carbohydrate lowfat diet. This increase in daylong glucose concentrations
occurred, despite the fact that insulin dose was significantly increased ( + 16%, P < 0.05) after the increase
in dietary carbohydrate to maintain fasting plasma glucose concentrations.
The results of these studies in individuals with diabetes are remarkably similar and consistent with the
overall findings in nondiabetic individuals. They indi-
777
GLUCOSE AND LIPOPROTEIN METABOLISM IN DIABETES
Total
N.S.
VLDL
p<0.05
LDL
N.S.
HDL
p<0.02
emerge, including increased fasting and postprandial triglyceride concentrations, increased postprandial glucose and insulin concentrations, and a decrease in HDL
cholesterol concentrations (Table 1). Moreover, improvements in fasting total plasma and LDL cholesterol
are not consistently demonstrated. When a reduction in
LDL cholesterol has been observed, it has been accompanied by a significnt increase in VLDL cholesterol and
VLDL triglyceride (8,13). Given the observed relationship between VLDL and LDL cholesterol and VLDL triglyceride and HDL cholesterol, one must be cautious in
interpreting the clinical significance of small reductions
in LDL cholesterol under these conditions.
200r_L
150
T
100
50
• 4 0 % CHO Diet
0 6 0 % CHO Diet
FIG. 3. Example of changes in fasting plasma lipoprotein
cholesterol concentrations that have been observed in individuals with non-insulin-dependent diabetes after diets
containing either 40 or 60% of total calories as carbohydrate (CHO). VLDL, very-low-density lipoprotein; LDL,
low-density lipoprotein; HDL, high-density lipoprotein.
From Coulston et al. (7). © by the American Journal of
Medicine.
cate that when carbohydrate is substituted for saturated
fat in the diet a constellation of deleterious metabolic
effects to both carbohydrate and lipoprotein metabolism
HIGH-CARBOHYDRATE MODIFIED-FAT
CHOLESTEROL-RESTRICTED DIETS
From the discussion of the literature thus far, it appears
that the putative beneficial effects on plasma lipoprotein
cholesterol concentrations, for which low-fat diets have
been recommended, have not been observed. However, not all high-carbohydrate low-fat diets have lead
to the same conclusions. In fact, there are several controlled metabolic studies that report significant improvements in both carbohydrate and lipoprotein metabolism
200
(D
O
O
150
(0 100
)
50
10
_
8
10
12
2
4
8
10
12
2
4
Time (Hrs)
60% CHO
40% CHO
FIG. 4. Fasting and postprandial plasma glucose (A) and insulin (B) concentrations observed in individuals with noninsulin-dependent diabetes after diets containing either 40% (•) or 60% (O) of total calories as carbohydrate (CHO).
From Coulston et al. (8). © by the American Diabetes Association.
778
DIABETES CARE, VOL. 14, NO. 9, SEPTEMBER 1991
C.B. HOLLENBECK AND A.M. COULSTON
TABLE 1
Metabolic effects of isocaloric substitution of carbohydrate for fat in diets of individuals with non-insulin-dependent
diabetes
Refs.
Fasting plasma
Total cholesterol
VLDL cholesterol
LDL cholesterol
HDL cholesterol
Total triglyceride
VLDL triglyceride
Postprandial plasma
Clucose
Insulin
Triglyceride
Sestoft
etal. (13)
Coulston
et al. (7)
Coulston
et al. (8)
Garg
etal. (14)
Rivellese
etal. (15)
NS
NS
NS
NS
NS
NS
i
1
t
T
t
NS
t
1
t
t
t
1
1
t
T
f
T
t
T
t
T
t
t
NS
1
T
*
T
T
T
r
NS
NS
T
T
NS
NS
*
NS, no significant change; VLDL, very-low-density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein. | , Increase; |
decrease.
*No data available.
in individuals with diabetes after high-carbohydrate
low-fat diets (10,16,19,21,22). For example, Hollenbeck et al. (10) reported significant improvements in
fasting total plasma (201 vs. 156 mg/dl, P < 0.05) and
LDL cholesterol (126 vs. 90 mg/dl, P < 0.05) concentrations in young individuals with IDDM after 6 wk of
high-carbohydrate low-fat cholesterol-restricted diets.
On the other hand, the low-fat diet resulted in higher
fasting plasma (115 vs. 96 mg/dl, P < 0.01) and VLDL
triglyceride (76 vs. 59 mg/dl, P < 0.001) concentrations
and a significant decrease in HDL cholesterol concentrations (50 vs. 39 mg/dl, P < 0.05). There were no
changes in either plasma glucose concentrations or insulin dose between the two diet periods. Although this
study demonstrated improvement in lipoprotein metabolism associated with the high-carbohydrate low-fat
diet, the improvements were limited to a fall in total
plasma and LDL cholesterol and are difficult to interpret
in the light of the concomitant increase in total plasma
and VLDL triglyceride and significant fall in HDL cholesterol concentrations. It is also important to understand that the diets used by these investigators did not
represent the isocaloric substitution of carbohydrate for
fat. The polyunsaturated-saturated ratio (0.14 vs. 1.40)
was significantly higher and dietary cholesterol significantly lower (580 vs. 62 mg/day) on the high-carbohydrate low-fat diet. One must consider the nature of
the metabolic changes observed in this study in relation
to the dietary changes. For example, based on current
evidence one could argue that the improvements in total
plasma and LDL cholesterol observed in this study resulted primarily from the changes in the polyunsaturated-saturated ratio and decrease in dietary cholesterol,
whereas the deleterious effects to total plasma and VLDL
triglyceride and HDL cholesterol resulted from increased carbohydrate. This will be discussed in more
detail below.
DIABETES CARE, V O L. 14, N O . 9, SEPTEMBER 1991
Abbott et al. (19) compared baseline diets containing
45% carbohydrate, 43% fat with a polyunsaturated-saturated ratio of 0.3, and 9 g fiber/1000 kcal with highcarbohydrate low-fat diets containing 65% carbohydrate, 21% fat with a polyunsaturated-saturated ratio of
1.2, and 18 g fiber/1000 kcal in seven Pima Indians
with NIDDM. The results demonstrated no significant
changes in total plasma or HDL cholesterol, total plasma
triglyceride, fasting plasma glucose, or insulin concentrations. On the other hand, there was a significant lowering in LDL cholesterol concentrations (107 vs. 84
mg/dl, P < 0.01). At first glance, these data also appear
to be in contrast to several previous studies comparing
the metabolic effects of high-carbohydrate diets
(7,8,13-15). Again, if one considers the results in the
light of their experimental dietary protocols, these differences become more apparent than real. For example,
a fall in LDL cholesterol concentration almost always
follows the reduction of saturated fat, regardless of
whether it is replaced by carbohydrate (10-12,16,20)
or unsaturated fats (11,12,18,20). The observation that
the reduction of total fat in the diet did not lead to any
significant fall in total plasma cholesterol is also an almost uniform finding (5-8,10,12-17,19,20). Finally,
the fact that increased dietary carbohydrate did not lead
to the expected deleterious effects on plasma triglyceride, glucose, and insulin concentration may be explained primarily by the judicious choice of dietary
carbohydrate and fiber. Specifically, these investigators
chose to increase carbohydrate by including sources of
legumes (pinto, kidney, and lima beans, lentils, and split
peas). As a result, the concomitant increase in dietary
fiber was primarily associated with an increase in leguminous fiber. There is evidence that legumes in general
(16,54,55), and perhaps leguminous fiber more specifically (16,55,56), can attenuate the deleterious metabolic effects of high-carbohydrate diets.
779
GLUCOSE AND LIPOPROTEIN METABOLISM IN DIABETES
At this juncture, it would seem important to stress that
the differences in the literature attributed to amount of
fat and carbohydrate may well result from differences in
experimental approach rather than any inherent inconsistency in results per se. For example, our group (5.-8)
as well as other investigators (11-15) have tried to assess
the role of lowering dietary fat on various aspects of
carbohydrate and lipoprotein metabolism. In these studies, total fat was replaced by either carbohydrate or
other unsaturated fats in such a manner that the polyunsaturated-saturated ratio, dietary cholesterol, dietary
fiber, and other factors known to affect carbohydrate and
lipid metabolism remained constant between the two
experimental diets. This is critical in understanding the
apparent differences reported in the literature in regard
to the effects of low-fat diets on various aspects of carbohydrate and lipoprotein metabolism. It appears that
virtually all controlled metabolic studies that report improvements in lipoprotein metabolism after high-carbohydrate low-fat diets (19,21-23) have not isolated
changes in the relative amount of fat from those due to
variations in either the kind of fat (saturated/monounsaturated/polyunsaturated) or amount of dietary cholesterol. Because dramatic changes in the relative proportions of saturated to polyunsaturated fat and
decreases in the amount of dietary cholesterol alone
could explain the observed improvements in plasma
lipid concentrations, it is not clear that the lower plasma
cholesterol concentrations could be attributed solely to
a reduction in dietary fat (57-59). Moreover, the relative
reduction of saturated fat has also been shown to have
a profound effect on plasma lipoprotein metabolism,
independent of dietary cholesterol content (57,58). Insight into the importance of modifying the composition
of dietary fat and the restriction of dietary cholesterol in
controlling plasma cholesterol concentrations is provided by two earlier studies in individuals with diabetes
(17,18). In one study, Weinsier et al. (17) evaluated the
effects of replacing total fat with carbohydrate with diets
in which the polyunsaturated-saturated ratio and dietary
cholesterol remained constant. These investigators
failed to demonstrate any significant differences in total
plasma cholesterol in 18 individuals with NIDDM after
20 wk of diets containing 60% of the calories as carbohydrate compared with conventional 40%-carbohydrate diets. With an entirely different approach, Kaufmann et al. (18) studied 270 individuals with IDDM on
diets with similar carbohydrate (40%) and fat (40%) content, but differing in the polyunsaturated-saturated ratio
(0.1 vs. 1.0) and amount of dietary cholesterol (7001500 vs. 300 mg/day). Their results showed a substantial lowering in total plasma cholesterol over the 11- to
17-day periods that the modified-fat cholesterol-restricted diets were fed. Thus, it appears likely that the
improvements in plasma cholesterol concentrations,
which were reported by several investigators (10,16,
19,21-23) to occur in individuals after the use of lowfat diets, can be attributed to the concomitant changes
in the composition of dietary fat and amount of dietary
780
cholesterol rather than to any specific effect of decreased fat.
Finally, it is now clear that specific groups of carbohydrate-containing foods (most prominently legumes)
may delay digestion and absorption, and thus result
in lower postprandial glucose and insulin response
(16,54-56). The relatively unique ability of leguminous
carbohydrate to attenuate the hyperglycemia, hyperinsulinemia, and hypertriglyceridemia observed after increased dietary carbohydrate is almost certainly due to
the unique structure and/or physical form of the carbohydrate present (60-62). The attenuation in hyperglycemia, hyperinsulinemia, and hypertriglyceridemia
that have been observed in studies utilizing leguminous
carbohydrate should not be attributed to the fact that
they are a complex carbohydrate. There is no evidence
that we are aware of which wojjld support the notion
that complex carbohydrate provides any advantage over
natural-occurring simple carbohydrate in controlling abnormalities of either carbohydrate or lipoprotein metabolism. Finally, the relatively unique effects of leguminous carbohydrate-containing foods should not
be generalized to other food sources rich in complex
carbohydrate. In other words, it is important that the
results obtained with large amounts of legumes as the
source of carbohydrate not be generalized to nonleguminous sources of carbohydrate.
UNSATURATED FATS: MONOUNSATURATED
VERSUS POLYUNSATURATED
Although, it seems to have become fashionable to emphasize the replacement of saturated fat with monounsaturated fats in recent studies, it is important to understand that the experimental approach is not novel. It
is essentially similar to the approach used by our group
(5-8) and others (13,16,18,19,21-23,63) which have
replaced saturated fat with either polyunsaturated fat or
mixtures of unsaturated fats. The notion that monounsaturated fat may provide metabolic advantage over
other unsaturated fats in this experimental paradigm is
questionable. For example, it has been argued that
monounsaturated fat lowers HDL cholesterol levels to
a lesser extent than polyunsaturated fat and therefore
may provide greater advantage when replacing saturated fat in the diet (64,65). Whether diets containing
predominantly polyunsaturated fat result in lower
HDL cholesterol compared with diets containing primarily monounsaturated fat is unclear (63-71). Several
studies report a decline in HDL cholesterol with the use
of polyunsaturated fat (64-66), whereas others have not
(63,67-77). Note that studies that report a decline in
HDL with the use of polyunsaturated fat have relied on
diets in which the polyunsaturated-saturated ratio was
high (2-6.5) (64-66), whereas those that demonstrate
comparable effects of polyunsaturated and monounsaturated fats have used somewhat more realistic polyun-
DIABETES CARE, VOL. 14, N O . 9, SEPTEMBER 1991
C.B. HOLLENBECK AND A.M. COULSTON
saturated-saturated ratios (0.6-1.2) (63,67-71). In
addition, it appears that both polyunsaturated and monounsaturated fats are equally effective in lowering total
and LDL cholesterol concentrations. Therefore, it seems
reasonable to conclude that replacing saturated fat with
unsaturated fat, whether it is primarily polyunsaturated,
monounsaturated, or mixtures of both, would result in
essentially similar metabolic outcomes in terms of lipoprotein metabolism.
EFFECTS OF INCREASED DIETARY CARBOHYDRATE
As mentioned earlier, an increase in total plasma and
VLDL triglyceride concentrations are a common finding
in individuals with diabetes. This is particularly true of
individuals with NIDDM (31-34). Concern for the presence of hypertriglyceridemia in individuals with diabetes is amplified by the strong relationship between
hypertriglyceridemia and the occurrence of CAD in this
population (41,42). Because there is extensive evidence
that low-fat high-carbohydrate diets lead to an increase
in plasma triglyceride concentrations (5-11,13-16,7274), this is an obvious drawback to the use of such diets
in individuals with diabetes. It has been argued that the
carbohydrate-induced hypertriglyceridemic effect of
low-fat high-carbohydrate diets can be attenuated if
such diets are greatly enriched with dietary fiber (16,
19,21,22). Although this may be true, the amount of
dietary fiber needed to accomplish this goal has generally been five to seven times the quantity of dietary
fiber that is consumed in the United States or Great
Britain and appears to be an amount that most individuals do not find acceptable (75,25). Moreover, it is now
apparent that only specific types of dietary fiber may be
successful in attenuating carbohydrate-induced hypertriglyceridemia (9,76,77). Therefore, the ability to prevent the deleterious effects of high-carbohydrate diets
would depend to a large extent on the ability to institute
the necessary modification in the type and amount of
dietary fiber. Because the studies that have attempted
these modifications have been largely unsuccessful
(75,25), it would seem reasonable to assume that adopting a high-carbohydrate low-fat diet would lead to increased triglyceride concentrations in most individuals
with diabetes. Indeed, when Anderson et al. (21) reduced the fiber content of his low-fat high-carbohydrate
high-fiber (34 g/1000 kcal) diets toward a more conventional level (12 g/1000 kcal), he noted the development of hypertriglyceridemia ( + 28%, P < 0.01) in
his group of individuals with NIDDM who had previously not shown this phenomenon of the high-fiber diet.
It must be emphasized that the fiber content of the lowfat high-carbohydrate diet that led to an increase in
plasma triglyceride concentrations was not low, but was
essentially identical to current dietary recommendations
(1). A study by our group was unable to demonstrate
any beneficial effect on either carbohydrate or lipopro-
DIABETES CARE, VOL. 14, N O . 9, SEPTEMBER 1991
tein metabolism when dietary fiber was increased from
11 to 27 g/1000 kcal in individuals with NIDDM consuming high-carbohydrate low-fat diets (9). In this
study, there was a significant increase in fasting plasma
triglyceride concentrations that occurred after the highcarbohydrate diet, which was unchanged by increasing
dietary fiber. Thus, it appears that hypertriglyceridemia
might be viewed as the likely consequence if individuals
with diabetes, particularly NIDDM, complied with the
current dietary advice to increase dietary carbohydrate.
It has also been argued that carbohydrate-induced
hypertriglyceridemia is a transient phenomenon. However, some studies that have reevaluated the hypertriglyceridemia associated with increased dietary carbohydrate do not support this conclusion (68,78). More
importantly, there appears to be a conspicuous absence
of any data in individuals with diabetes, which would
support an evanescent nature of carbohydrate-induced
hypertriglyceridemia. The available data from several
controlled metabolic studies have reported the persistence of carbohydrate-induced hypertriglyceridemia for
at least 6 wk (8,10,14). This is perhaps best illustrated
in a study by our group (8). Fasting plasma triglyceride
concentrations were significantly elevated by the end of
the 1st wk of the high-carbohydrate low-fat diet and
remained unchanged for the remainder of the 6-wk period (Fig. 5). The only long-term study of the effects of
low-fat high-carbohydrate diets on plasma triglyceride
concentrations is, unfortunately, difficult to interpret.
Stone and Connor (79) reported no significant increase
in plasma triglyceride in individuals with IDDM after 1
yr of a cholesterol-restricted low-fat high-carbohydrate
diet. These results have been interpreted as supporting
the view that carbohydrate-induced hypertriglyceridemia is a transitory phenomenon. However, fasting triglyceride concentrations were measured only at the
beginning and end of the 1 -yr study, and the concentrations of triglyceride was unchanged. Therefore, there
was no evidence presented to indicate that the dietary
changes resulted in an increase in triglyceride concentrations. The individuals studied by Stone and Connor
(79) had IDDM and their glycemia was well controlled.
Plasma triglyceride concentrations are rarely elevated in
such individuals (34,38). Consequently, these data
could also be interpreted as demonstrating that low-fat
high-carbohydrate diets do not necessarily lead to an
increase in plasma triglyceride concentrations in wellcontrolled individuals with IDDM.
Another potential deleterious effect, which has been
described in individuals with diabetes after high-carbohydrate low-fat diets, is a significant reduction in HDL
cholesterol concentrations. The observation that highcarbohydrate diets lead to significant reductions in HDL
cholesterol appears to be a common phenomenon and
has been shown to occur in nondiabetic individuals
(5,6,11,12) and in individuals with diabetes mellitus
(7,8,13-15). Because HDL cholesterol concentration
has been shown to be inversely associated with the development of CAD, it is difficult to easily dismiss the
781
GLUCOSE AND LIPOPROTEIN METABOLISM IN DIABETES
240
220
180
140
100
60
20
3
4
Weeks
60%
40%
CHO
CHO
FIG. 5. Weekly fasting plasma triglyceride concentrations
in individuals with non-insulin-dependent diabetes after
diets containing either 40% (•) or 60% (O) of total calories
as carbohydrate (CHO). Data indicate persistent nature of
CHO-induced hypertriglyceridemia observed after highCHO low-fat diets. From Coulston et al. (8). © by the
American Diabetes Association.
significance of the decrease in HDL cholesterol associated with high-carbohydrate diets (43-46).
Finally, there is the observation that increasing dietary
carbohydrate in individuals with diabetes leads to a
significant deterioration in carbohydrate metabolism
(7,8,13-15). Specifically, high-carbohydrate diets have
resulted in increased daylong glucose and insulin concentrations and increased urinary glucose excretion.
This appears to be a common finding in individuals with
diabetes, unless the major source of dietary carbohydrate contains relatively large amounts of legumes and
leguminous fiber (16,19,21,22).
Although high-carbohydrate diets are generally believed to increase insulin sensitivity (80-83), the experimental evidence in support of this notion is not as
convincing as generally perceived. First, most of the
studies cited in this regard do not actually measure insulin sensitivity directly and thus do not provide compelling evidence that the use of high-carbohydrate diets
enhance insulin action (80-82). There are relatively few
studies that have measured insulin action directly, and
the results of these appear to be mixed. In nondiabetic
782
glucose-tolerant individuals, Kolterman et al. (83) reported an increase in insulin-stimulated glucose disposal
rate after diets that contained 75% of total calories as
carbohydrate, whereas Borkman et al. (84) failed to
demonstrate any significant improvement in insulinstimulated glucose uptake after diets containing 55% of
calories as carbohydrate. It is possible that improvements in insulin action are observed only at the higher
carbohydrate intakes used by Kolterman et al. (83).
More importantly, however, the small improvements in
insulin action demonstrated by Kolterman et al. should
not divert attention from the fact that ambient plasma
glucose and insulin concentrations were significantly elevated after the high-carbohydrate diet. Thus, the small
improvements observed in insulin-stimulated glucose
disposal were apparently insufficient to compensate for
the hyperglycemic and hyperinsulinemic effects of the
increased carbohydrate in individuals with normal glucose tolerance. In the only study to measure insulin
action in individuals with diabetes, Garg et al. (85) reported no significant improvement in insulin action after
diets containing 60% of calories as carbohydrate. In addition, studies assessing the effects of high-fiber diets on
insulin action have been negative (86,87), despite speculation that high-fiber diets lead to improvements in insulin action (88). There is now substantial evidence that
suggests that the improvement in postprandial glucose
on high-fiber diets results from gastrointestinal effects
delaying digestion and absorption (88).
CONCLUSIONS
The major focus of dietary advice for individuals with
diabetes over the past several years has been the emphasis on high-carbohydrate low-fat diets. The rationale
for this approach appears to be based on the premise
that such diets result in a decrease in LDL cholesterol
concentrations and ultimately reduce the risk and incidence of CAD in individuals with diabetes. However,
replacing dietary fat, or more specifically saturated fat,
with dietary carbohydrate may not be the most appropriate means of reducing cardiovascular risk factors in
this population. As reviewed herein, there are conflicting reports of the metabolic impact of high-carbohydrate
low-fat diets. Much of the confusion seems to be related
to the differences in dietary protocols. After careful review, it seems essential to separate the changes due to
differences in the relative amount of fat and carbohydrate from those due to variations in the kind of fat and
carbohydrate and amount of dietary cholesterol. There
is now substantial evidence that suggests that diets containing conventional quantities of fat (—40% of total
calories), in which the composition of fat has been modified to reduce saturated fatty acids and dietary cholesterol, appear to offer the best overall control of
carbohydrate and lipoprotein metabolism in individuals
with diabetes. Data indicating that high-carbohydrate
low-fat diets lead to beneficial effects on carbohydrate
DIABETES CARE, V O L. 14, N O . 9, SEPTEMBER 1991
C.B. HOLLENBECK AND A.M. COULSTON
or lipoprotein metabolism are confounded either by the
lack of suitable experimental control, or by the fact that
diets also differed in the type of dietary fat and amount
of dietary cholesterol, or were enormously enriched in
dietary fiber. When these factors are taken into consideration, there appears to be little evidence in support of
the view that substituting carbohydrate for fat in the diets
of individuals with diabetes results in any measurable
beneficial effect on either carbohdyrate or lipoprotein
metabolism. Indeed, it could be argued that the available evidence supports the conclusion that the most
characteristic defects in carbohydrate and lipoprotein
metabolism in individuals with diabetes have a marked
tendency to deteriorate in response to increased dietary
carbohydrate. Consequently, we believe that the available evidence does not support the view that individuals
with diabetes should be encouraged to follow high-carbohydrate low-fat diets. This statement should not be
interpreted to indicate that dietary fat is unimportant.
On the contrary, in this review, we have attempted to
stress that the available data support the importance of
type of dietary fat and amount of dietary cholesterol in
the regulation of lipoprotein metabolism in individuals
with diabetes. The replacement of saturated fat with unsaturated fat and reduction of dietary cholesterol in the
context of conventional diets results in the desired fall
in total and LDL cholesterol without the deleterious effects of hyperglycemia, hyperinsulinemia, hypertriglyceridemia, and reduced HDL cholesterol concentrations.
ACKNOWLEDGMENTS
This work was supported in part by Human Health Services Grant M01-RR-00070, General Clinical Research
Centers, Division of Research Resources, National Institutes of Health.
REFERENCES
1. American Diabetes Association: Nutritional recommendations and principles for individuals with diabetes mellitus: 1986. Diabetes Care 10:126-32, 1987
2. Canadian Diabetes Association: Special Report Committee guidelines for the nutritional management of diabetes
mellitus: a special report from the Canadian Diabetes Association. / Can DietAssoc 42:110-18, 1981
3. Reaven GM: How high the carbohydrates? Diabetologia
19:409-13, 1980
4. Hollenbeck CB, Coulston AM, Reaven GM: Dietary carbohdyrate: a different perspective. Diabetes Care 9:64147, 1986
5. Coulston AM, Liu GC, Reaven GM: Plasma glucose, insulin, and lipid responses to high-carbohydrate low-fat
diets in normal humans. Metabolism 32:52-56, 1983
6. Liu GC, Coulston AM, Reaven GM: Effect of high-carbohydrate low-fat diets on plasma glucose, insulin and
lipid responses in hypertriglyceridemic humans. Metabolism 32:750-53, 1983
DIABETES CARE, V O L . 14, N O . 9, SEPTEMBER 1991
7. Coulston AM, Hollenbeck CB, Swislocki AML, Chen YDl, Reaven GM: Deleterious metabolic effects of highcarbohydrate, sucrose containing diets in patients with
NIDDM. Am / Med 82:213-320, 1987
8. Coulston AM, Hollenbeck CB, Swislocki AML, Reaven
GM: Persistence of hypertriglyceridemic effect of low-fat
high-carbohydrate diets in NIDDM patients. Diabetes
Care 12:94-101, 1989
9. Hollenbeck CB, Coulston AM, Reaven GM: To what extent does increased dietary fiber improve glucose and
lipid metabolism in patients with noninsulin-dependent
diabetes mellitus (NIDDM)? Am / Clin Nutr 43:16-24,
1984
10. Hollenbeck CB, Connor WE, Riddle MC, Alaupovic P,
Leklem JE: The effects of a high-carbohydrate, low-fat,
cholesterol restricted diet in plasma lipids, lipoprotein
and apoprotein concentration in insulin dependent (type
I) diabetes mellitus. Metabolism 34:559-66, 1985
11. Grundy SM: Comparison of monounsaturated fatty acid
and carbohydrate in lowering plasma cholesterol. N Engl
I Med 314:745-48, 1988
12. Grundy SM, Florentin L, Nix D, Whelan MF: Comparison
of monounsaturated fatty acids and carbohydrate in man.
Am J Clin Nutr 47:965-69, 1988
13. Sestoft L, Krarup T, Palmvig B, Meinertz H, Faergeman
O: High-carbohydrate, low-fat diet: effect on lipid and
carbohydrate metabolism, GIP and insulin secretion in
diabetics. Dan Med Bull 32:64-69, 1985
14. GargG, Bonanome A, Grundy SM, Zhang Z-J, Unger RH:
Comparison of a high-carbohydrate diet with a highmononunsaturated fat diet in patients with non-insulindependent diabetes mellitus. N Engl I Med 319:829-34,
1988
15. Rivellese A, Giacco R, Genovese S, Patti L, Marotta G,
Pacioni D, Annuzzi G, Riccardi G: Effects of changing
amount of carbohydrate in diet on plasma lipoproteins
and apolipoproteins in type II diabetic patients. Diabetes
Care 13:446-48, 1990
16. Riccardi G, Rivellese A, Pacioni D, Genovese S, Mastranzo P, Mancini M: Separate influence of dietary carbohydrate and fiber on metabolic control of diabetes.
Diabetologia 26:116-21, 1984
17. Weinsier RL, Seeman A, Herrera MG: High- and lowcarbohydrate diet in diabetes mellitus: study of effects on
diabetic control, insulin secretion, and blood lipids. Ann
Intern Med 80:332-41, 1974
18. Kaufmann RL, Assayl JP, Soeldner JS: Plasma lipid levels
in diabetic children: effect of diet restricted in cholesterol
and saturated fats. Diabetes 24:672-79, 1975
19. Abbot WGH, Boyce VL, Grundy SM, Howard BV: Effects
of replacing saturated fat with complex carbohydrate in
diets of subjects with NIDDM. Diabetes Care 12:102107, 1989
20. Lewis B, Hammett F, Katan M: Toward an improved lipidlowering diet: additive effects of changes in nutrient intake. Lancet 2:1310-313, 1981
21. Anderson JW, Chen WJL, Sieling B: Hypolipemic effects
of high-carbohydrate high-fiber diets. Metabolism 29:
551-58, 1980
22. Kiehm TG, Anderson JW, Ward K: Beneficial effects of a
high-carbohydrate, high fiber diet on hyperglycemic diabetic men. Am / Clin Nutr 29:895-99, 1976
23. Abbott WGH, Swinburn B, Giacomo R, Hara H, Patti L,
Harper I, Grundy SM, Howard BV: Effect of a high-carbohydrate, low-saturated-fat diet on apolipoprotein B and
783
GLUCOSE AND LIPOPROTEIN METABOLISM IN DIABETES
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
triglyceride metabolism in Pima Indians. / Clin Invest
86:642-50, 1990
Simpson HCR, Simpson RW, Lousley S: High carbohydrate leguminous fibre diet improves all aspects of dietary
control. Lancet 1:1-5, 1981
Simpson HCR, Simpson RW, Lousley S: Digestible carbohydrate: an independent effect on diabetic control in
type 2 (non-insulin-dependent) diabetic patients? Diabetologia 23:225-39, 1982
Simpson RW, Mann Jl, Eaton J: Improved glucose control
in maturity-onset diabetes treated with high-carbohydrate-modified fat diets. Br j Med 1:1753-56, 1979
Hayford JT, Thompson RG: Free and total insulin integrated concentrations in insulin dependent diabetes. Metabolism 31:387-97, 1982
Reaven GM, Chen Y-DI, Coulston AM, Greenfield MA,
Hollenbeck CB, Lardinois C, Liu GC, Schwartz H: Insulin
secretion and insulin action in noninsulin-dependent diabetes mellitus: is insulin resistance secondary to hyperinsulinemia? Am I Med 75:85-93, 1983
Liu GC, Coulston AM, Chen Y-DI, Reaven GM: Does
daylong absolute hypoinsulinemia characterize the patient with non-insulin-dependent diabetes mellitus? Metabolism 32:754-56, 1983
Coulston AM, Hollenbeck CB, Reaven GM: The effect of
source of dietary carbohydrate on plasma glucose and
insulin responses to mixed meals in subjects with noninsulin-dependent diabetes mellitus. Diabetes Care 10:
395-400, 1987
Bagdade JD, Porte D, Bierman EL: Diabetic lipemia: form
of acquired fat involved lipemia. N Engl) Med 276:42733, 1967
Nikkila EA, Kekki M: Plasma triglyceride transport kinetics in diabetes mellitus. Metabolism 22:1-22, 1973
Kissebah AH, Adams PW, Wynn V: Inter-relationship between insulin secretion and plasma free fatty acids and
triglyceride transport kinetics in maturity onset diabetes
and the effect of phenephylibiguanide (phenformin). Diabetologia 18:119-30, 1974
Greenfield M, Kolterman O, Olefsky JM, Reaven GM:
Mechanism of hypertriglyceridemia in diabetic patients
with fasting hypertriglyceridemia. Diabetologia 18:44146, 1980
Ginsberg H, Grundy SM: Very low density lipoprotein
metabolism in non-ketotic diabetes mellitus: effect of dietary restriction. Diabetologia 23:421-25, 1982
Nikkila EA: High-density lipoproteins in diabetes. Diabetes 30 (Suppl. 2):82-87, 1984
Hollenbeck CB, Chen Y-DI, Greenfield MS, Lardinois CK,
Reaven GM: Reduced plasma high density lipoprotein
cholesterol concentrations need not increase when hyperglycemia is controlled with insulin in non-insulin dependent diabetes mellitus (NIDDM). ) Clin Endocrinol
Metab 62:605-608, 1986
Nikkila EA: Plasma lipid and lipoprotein abnormalities in
diabetes. In Diabetes and Heart Disease. Jarett RJ, Ed.
New York, Elsevier, 1984, p. 133-67
39. Carlson LA, Bottiger LE, Ahfeldt PE: Risk factors for myocardial infarction in the Stockholm perspective study: a
14 year follow-up focusing on the role of plasma triglyceride and cholesterol. Ada Med Scand 706:351-60,
1979
40. Santen RJ, Willis PW, Fajans SS: Atherosclerosis in diabetes mellitus: correlation with serum lipids, adiposity,
and serum insulin levels. Arch Intern Med 130:
784
833-43, 1972
41. West KM, Ahuja MMS, Bennett PH, Czyzyk A, DeAcosta
OM, Fuller JH, Grab B, Grabauskas V, Jarrett RS, Kosaka
K, Keen H, Krolewski AS, Miki E, Schliack V, Teuscher
A, Watkins PJ, Stober JA: The role of circulating glucose
and triglyceride concentrations and their interaction with
other "risk factors" as determinants of arterial disease in
nine diabetic populations: samples from the WHO Multinational study. Diabetes Care 6:361-69, 1983
42. Fontbonne A, Eschwege E, Cambien F, Richard JL, Ducimetiere P, Thibult N, Warnet JM, Claude JR, Rosselin
GE: Hypertriglyceridemia as a risk factor of coronary heart
disease mortality in subjects with impaired glucose tolerance or diabetes: results from the 11-year follow-up of
the Paris Prospective Study. Diabetologia 32:300-304,
1989
43. Miller GJ, Miller NE: Plasma high-density lipoprotein
concentration and development of ischaemic heart disease. Lancet 1:16-19, 1975
44. Castelli WP, Doyle JT, Gordon T: HDL cholesterol and
other lipids in coronary heart disease: the cooperative
lipoprotein phenotype study. Circulation 55:767-72,
1977
45. Reckless JPD, Betterridge DJ, Wu P, Payne B, Galton DJ:
High-density and low-density lipoproteins and prevalence of vascular disease in diabetes mellitus. Br Med }
1:883-86, 1978
46. Beach KW, Brunzell JD, Conquest LL, Strandness DE: The
correlation of atherosclerosis obliterans with lipoproteins
in insulin-dependent and non-insulin-dependent diabetes. Diabetes 28:836-40, 1979
47. Havel RJ: High density lipoproteins, cholesterol transport
and coronary heart disease. Circulation 60:1-3, 1979
48. Nestel PJ, Readon M, Billington T: In vivo transport of
cholesteryl esters from high density lipoproteins to very
low density lipoproteins in man. Biochem Biophys Acta
573:403-407, 1979
49. Stout RW, Vallance-Owen J: Insulin and atheroma. Lancet
1:1078-80, 1969
50. Zavaroni I, Bonora E, Pagliara M, Dall'Aglio E, Luchetti
L, Buonanna G, Bonati PA, Bergonzani M, Gnudi L, Passeri M, Reaven GM: Risk factors for coronary artery
disease in healthy persons with hyperinsulinemia and
normal glucose tolerance. N Engl I Med 320:702-706,
1989
51. Hamsten A, Efendic S, Wallsius, Szamosi A, de Faire V:
Glucose tolerance and insulin response to glucose in nondiabetic young male survivors of myocardial infarction.
Am Heart} 113:917-27, 1987
52. Kuo PT, Feng LY: Study of serum insulin in atherosclerotic
patients with endogenous hypertriglyceridemia (types III
and IV hyperlipoproteinemia). Metabolism 19:372-80,
1970
53. Stout RW: Insulin and atheroma: 20-yr perspective. Diabetes Care 13:631-54, 1990
54. Jenkins DJA, Wolever TMS, Taylor RH, Barker HM, Fielden H: Exceptionally low blood glucose response to dried
beans: comparison with other carbohydrate foods. Br
Med I 2:578-80, 1980
55. Jenkins DJA: Leguminous seeds in the dietary management of hyperlipidemia. Am I Clin Nutr 38:567-73, 1983
56. Jenkins DJA, Leeds AR, Gassull MA, Cochet B, Alberti
KGMM: Decreased postprandial insulin and glucose concentrations for guar and pectin. Ann Intern Med 86:20—
23, 1977
DIABETES CARE, V O L . 14, N O . 9, SEPTEMBER 1991
C.B. HOLLENBECK AND A.M. COULSTON
57. Aherns EH, Hirsch J, Insull W: The influence of dietary
fats on serum lipid levels in man. Lancet 1:943-53, 1957
58. Connor WE, Stone DB, Hodges RE: The interrelated effects of dietary cholesterol and fat upon the human serum
lipid levels. J Clin Invest 43:1691-96, 1964
59. Connor WE, Hodges RE, Bleiler RE: The serum lipids in
men receiving high cholesterol and cholesterol-free diets.
I Clin Invest 40:894-901, 1961
60. Golay A, Coulston AM, Hollenbeck CB, Kaiser LL,
Wursch P, Reaven GM: Comparison of metabolic effects
of white beans processed into two different physical
forms. Diabetes Care 9:260-66, 1986
61. Wong S, O'Dea K: Importance of physical form rather
than viscosity in determining the rate of starch hydrolysis
in legumes. Am j Clin Nutr 37:66-70, 1983
62. Wursch P, Del Vedovo S, Kowllreutter B: Cell structure
and starch nature as key determinants of the digestion rate
of starch in legumes. Am I Clin Nutr 43:25-29, 1986
63. Mensink RP, Katan MB: Effect of a diet enriched with
monounsaturated or polyunsaturated fatty acids on levels
of low-density and high-density lipoprotein cholesterol in
healthy women and men. N Engl I Med 321:436-41,
1989
64. Mattson FH, Grundy SM: Comparison of effects of dietary
saturated, monounsaturated, and polyunsaturated fatty
acids on plasma lipids and lipoproteins in man. I Lipid
Res 26:194-202, 1985
65. Schaefer EJ, Levy Rl, Ernst DN, VanSant FD, Brewer HB
Jr: The effects of low cholesterol, high polyunsaturated
fat, and low fat diets on plasma lipid and lipoprotein cholesterol levels in normal and hypercholesterolemic subjects. Am / Clin Nutr 34:1758-63, 1981
66. Vega GL, Croszek E, Wolf R, Grundy SM: Influence of
polyunsaturated fats on composition of plasma lipoprotein and apolipoproteins. j Lipid Res 23:811-22, 1982
67. Mensink RP, de Groot MJM, van den Broeke LT, Severijnen-Nobels AP, Demacker PNM, Katan MB: Effect of
monounsaturated fatty acids vs. complex carbohydrates
on serum lipoproteins and apoproteins in healthy men
and women. Metabolism 38:172-78, 1989
68. Brussaard JH, Dallinga-Thie G, Groot PH, Katan MB: Effects of amount and type of dietary fat on serum lipids,
lipoproteins and apoproteins in man: a controlled 8-week
trial. Atherosclerosis 36:515-27, 1980
69. Schonfeld G, Patsch W, Rudel LL, Nelson C, Epstein M,
Olson RE: Effects of dietary cholesterol and fatty acids on
plasma lipoproteins. I Clin Invest 69:1072-80, 1982
70. Weisweiler P, Janetschek P, Schwandt P: Influence of
polyunsaturated fats and fat restriction on serum lipoproteins in humans. Metabolism 34:83-87, 1985
71. Mensink RP, Katan MB: Effect of monounsaturated fatty
acids versus complex carbohydrates on high-density lipoproteins in healthy men and women. Lancet 1:12225, 1987
72. Ginsberg H, Olefsky JM, Kimmerling G: Induction of hypertriglyceridemia by a low-fat diet.) Clin Endocrinol Me-
DIABETES CARE, V O L. 14, N O . 9, SEPTEMBER 1991
tab 42:729-35, 1976
73. Bierman EL, Hamlin JT: The hyperlipemic effect of lowfat, high-carbohydrate diet in diabetic subjects. Diabetes
10:432-37, 1961
74. Farquhar JW, Frank A, Gross RC, Reaven GM: Glucose,
insulin and triglyceride responses to high and low carbohydrate diets in man. ) Clin Invest 45:1648-56, 1966
75. Hollenbeck CB, Leklem JE, Riddle MC, Connor WE: The
composition and nutritional adequacy of subject selected
high-carbohydrate, low-fat diets in insulin dependent diabetes. Am ) Clin Nutr 38:41-51, 1983
76. Gatti E, Catenazzo G, Camisaach E: Effects of guar-enriched pasta on the treatment of diabetes and hyperlipidemia. Ann Nutr Metab 28:1-10, 1984
77. Munoz JM, Sandstead HH, Jacob CA: Effects of some
cereal brans and textured vegetable protein on plasma
lipids. Am ) Clin Nutr 32:580-92, 1979
78. Brussaard JH, Dallinga-Thie G, Goot PHE: Serum lipoproteins of healthy persons fed a low-fat diet or a polyunsaturated fat diet for three months. Atherosclerosis
42:205-19, 1982
79. Stone, D, Connor WE: The prolonged effects of a lowcholesterol, high-carbohydrate diet upon serum lipids in
diabetic patients. Diabetes 12:127-32, 1963
80. Brunzell JD, Lemer RL, Hazzard WR, Porte D, Bierman
EL: Improved glucose tolerance with high carbohydrate
feeding in mild diabetes. N Engl J Med 284:521-24, 1971
81. Anderson JW, Herman RH, Zakim D: Effect of high glucose and high sucrose diets on glucose tolerance of normal men. Am J Clin Nutr 26:600-607, 1977
82. Anderson JW: Effect of carbohydrate restriction and high
carbohydrate diets on men with chemical diabetes. Am )
Clin Nutr 30:402-408, 1977
83. Kolterman OG, Greenfield MS, Reaven GM, Saekow M,
Olefsky JM: Effect of a high-carbohydrate diet on insulinbinding to adipocytes and on insulin action in vivo in
man. Diabetes 28:731-36, 1979
84. Borkman M, Campbell LV, Chisholm DJ, Storlien LH:
Comparison of the effects on insulin sensitivity of high
carbohydrate and high fat diets in normal subjects. / Clin
Endocrinol Metab 72:432-37, 1991
85. Garg A, Grundy SM, Helderman JH, Unger RH: Highcarbohydrate diets do not induce insulin sensitivity in
NIDDM patients (Abstract). Diabetes 38 (Suppl. 2):46A,
1989
86. Nestel PJ, Nolan C, Bazelmans J, Cook R: Effects of
a high-starch diet with low- or high-fiber content on
postabsorptive glucose utilization and glucose production in normal subjects. Diabetes Care 7:207-10,
1984
87. Hoffman CR, Fineberg SE, Howey DC, Clark CM, Pronsky
Z: Short-term effects of a high-fiber high-carbohydrate
diet in very obese diabetic individuals. Diabetes Care
5:605-11, 1982
88. Vinik Al, Jenkins DJA: Dietary management of diabetes.
Diabetes Care 11:160-73, 1988
785